Akademik Veri Yönetim Sistemi
Journal of Mechanical Science and Technology, cilt.38, sa.3, ss.1227-1236, 2024 (SCI-Expanded)
This study focuses on the optimization of the bonding conditions for adhesive joints, specifically targeting PA66/GF and epoxy/CF thermoset materials. The Box-Behnken (BB) design was employed for meticulous optimization, considering variables such as mixing ratio (M1), curing time (M2), and primer application (M3). The quadratic model derived from BB demonstrated synergistic effects between variables, allowing for an accurate prediction of response strength. Ultrasonic testing has emerged as a powerful nondestructive evaluation (NDE) tool for assessing bond quality in thermoset materials. PA66/GF and epoxy/CF specimens underwent thorough examination, emphasizing the advantages of non-destructive techniques over traditional destructive tests. The results of the destructive lap shear tests provide a comprehensive understanding of the bonding conditions. The control specimen exhibited a load of 23.7 MPa, while the S-5 group of specimens with modified bonding conditions exhibited higher maximum loads (24.9 MPa), and the S-25 group exhibited a higher maximum loads of 25.6 MPa. The results obtained in this study effectively separated the samples with different adhesion conditions from the control sample. The developed model for the variation of bonding conditions exhibited high prediction accuracy, supported by significant F-values from the ANOVA tests. The regression coefficient (R-square value) of 0.9557 underlines the strong correlation between the variables, with M3 and M2 identified as the key factors. Correlating ultrasonic signals with shear load reveals a strong predictive relationship, with ratios of 0.94 (PA66/GF-adhesive interface) and 0.96 (Epoxy/CF-adhesive interfaces). The model effectively discerns the need for epoxy/CF adhesive interfaces, highlighting potential challenges for the BB-based approach. As the study concludes, the abstract encapsulates the optimization, evaluation, and predictive capabilities of the proposed model for adhesive bonding conditions.